Create a streamer for updating a linear regression model

LinearRegressions creates a streaming algorithm that can calculate and update linear regression coefficients.

Format

An R6Class generator object

Methods

Public methods

• LinearRegression$new()

• LinearRegression$update()

• LinearRegression$clone()

---

Method new()

Creates a new LinearRegression streamer object.

Usage

LinearRegression$new(X, y, fit_intercept = TRUE, method = "woodbury")

Arguments

X

initial design matrix

y

initial labels

fit_intercept

if `TRUE`, add an intercept to the model

method

the method used to update regression coefficients

Returns

The new LinearRegression (invisibly)

---

Method update()

Updates the regression coefficient based on new data

Usage

LinearRegression$update(X, y)

Arguments

X

new design matrix

y

new labels

#' @return The updated LinearRegression (invisibly)

---

Method clone()

The objects of this class are cloneable with this method.

Usage

LinearRegression$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

X <- matrix(c(3, 1, 4, 1, 5, 9, 2, 6), ncol = 2)
y <- c(2, 7, 1, 8)
reg <- LinearRegression$new(X, y)
X_new <- matrix(c(5, 3, 5, 8), ncol = 2)
y_new <- c(2, 8)
reg$update(X_new, y_new)